Topics‎ > ‎

4. Evaluation Model

 Car Ballooning studying buoys drift (tail)  sample image for map stitching - aerial photography - IMG_8114

protei --- future :)

P1060487 P1060491
We need to be able to evaluate the performances of Protei.
How close to the wind can it go (upwind sailing)?
How much pulling power? How many kg of payload? How does it affect maneuverability?
How quickly can it turn?
How are the interactions of the tail and the Oil?
What sensors are required to evaluate Protei? How much on-board? How much outside (observer)?
If we observe from 
    - Above? Protei 005 took place in a baby swimming pool. The robot was tagged, filmed with a webcam.
        It is easy to think of a helium balloon flying far above Protei, analyzing its movement. 
    - Below? Is it easier to have an object we drag from the bottom?
Do we need constant evaluation / do we need evaluation just for development time?
Do we need constant evaluation to avoid tail entanglement? How useful would that be? How much would it cost? How reliable?

<Towing tank testing - with Dr Gonzalo Tampier, UACH>

*Towing tank on the Hull
Can only measure the resistance force

model: 1m long
Challenge: drift forces
-->Rotate the model slightly and measure the resistance again

Test Matrix:

  1. 3 velocities
  2. 3 drift angles
  3. 3 heeling angles
    = 27 cases
    = extensive testing

-->Develop a Velocity Prediction Program using the measurements as input

*Towing tank on the boom

  1. drag of different lengths of booms
  2. drag booms with different weight/section
  3. ?? how to simulate oil 
  4. drag resistance in front waves

--> model scale of the boom
    = need scaling laws for the viscosity of oil:  ask TU Berlin, SINTEF, CEDRE

</Towing tank testing - with Dr Gonzalo Tampier, UACH>

<Summer testing campaign proposal for 2012>
How can we measure the different inputs for the VPP by using the vehicles we have in our possessions (or the one we can borrow) while having fun and providing great team building exercises?
The main challenge are: 
1. Source the parts
2. Install, plug, test, verify, calibrate
3. Collect data
4. Analyse data
</Summer testing campaign proposal for 2012>

<Summer testing campaign  - August 2011>
Suggested Planning for sailing testing team Roberto, Fiona & François & ??
  • Review tests and set up planning: Fiona + Roberto + Etienne
  • Heeling angle measurement?
  • Sit down to perform a Risk assessment: what can go wrong during testing? What if...?
  • Let De Kaag people know about your plans.
  • Check with Javier and Gonzalo the data they need to reproduce the tests in the VPP
  • Buy and install all needed equipment
  • Continue installing equipment
  • Start dry tests for Camera, Wind anemometre, GPS data exchange (try with a bike)
  • Get weather forecast + tide times and currents info
  • Work on boom ballasting
  • Work out the hull bending in sea water
  • Try out boom ballasting from the pier with a rope
  • Briefing before testing: need
    • Head of test controlling the operations from the pier/a mothership closeby testing area. 
    • 3 people in the sailing boat following Protei: 2 for maneuvres, 1 for RC. Need a forth one for wind measurement
    • 1 wind measurer in the pier/ mothership closeby
*Saturday and Sunday: 
Top priority is TEST 0 to prepare for TEST C1.

I think that´s more than enough for the few days coming. Forget about the tests A1, A2, B1 and B2. Time flies in testing. Don´t overload and prepare well your troops.

Test name





 0. Getting used to controlling  Protei Using RC we need to be able to keep a relatively constant course with Protei. Difficult to do because the response time of the hull is slow.
-Try to maintain a course that oscillates between close hauled and a beam reach (a close reach ideally)
-Then try to sail at other points of sail
-Keep a count of many times you accidentally tack/jibe/get stuck in irons
-Try to do some tacking/jibing (jibing not too important)
-Add oil boom and see how everything changes
 Entirely qualitative, pics videos etc
  • motor boat
  • camera
  • Walky Talkies

A1. Global drag

Pull rigid hull in straight line and measure the tension with the spring behind a motor boat. Towing Protei:

1. BEST:set up a large transverse beam across the motor boat and attach Protei on it. Position it so that you have as little wave interference as possible. Also convenient for filming. Observe how the hull naturally deforms without pulling any bending cable. Does the hull naturally bends when the motorboat turns? Is it possible to measure the natural bending by knowing the position of the cable / position of the cross section respective to each other?

2. OK: as far as possible from the motor, measure the tension at different speeds and report on a curve (influenced by the wake of motor boat)

Sketch of testing configuration

Testing configuration : Protei is dragged from the side of the boat and a camera attached on top to observe bending

indicative value in KG of tension at different speeds. Should increase with speed. 

Observe natural bending of hull

Speeds: 1, 2 and 3 knots.

  • spring scales
  • motor boat, boat with a mast
  • speedometer to measure boat speed
  • large transverse beam
  • camera set up
  • fixation for transverse beam

Saturday @ De Kaag sailing club

A2. Bending drag

Bend the hull to its maximum and define two intermediate bending positions that you can easily reach again - for example counting the number of actuator rotation/ length of cable pulled. Observe how Protei reacts to the bending: 

moving side ways? which side for which bending ? how much rudder angle do you need in the motorboat to align it to Protei again?

increase in resistance? how much for how much bending (1-2-3)?

indicative value in KG of tension at 3 speeds (1, 2 and 3 knots) and 3 bendings

  • spring scales
  • motor boat
  • boat speedometer

Saturday @ De Kaag sailing club

Prepare bending positions 1,2 and 3 in water beforehand. 

B1. VPP sailing speed

Measure the sailing speed at different wind speeds and course into the wind. Make sure that you measure the speed once the course is stable, not the transient speed when trimming the sail. --> SKIP TO OVERALL PERFORMANCE C1: quite a lot of polar diagrams points can be measured when sailing.

Polar diagram : sailing speed at different wind speeds and courses

  • GPS logger (Protei speed/course)
  • anemometer


B2. VPP heeling angle

Measure the heeling angle  at different wind speeds and course into the wind.

Estimate heeling angle from eyesight (<-- not really feasible!). Keep a checklist of which points of sail are stable (stable as in they don't cause the masthead to fall into the water) --> we need a clever way to measure heel angle: need to measure the heigh of the mast from a fixed point (for example water level). An altimeter on top of the mast? Then we need to perform a dry test to calibrate the heeling angle measure.

Polar diagram : heeling angle at differend wind speeds and courses

  • GPS logger (Protei speed/course)
  • anemometer


C1. Overall performance (Race Track)

Set up a race track with 3 buoys.

Go upwind in a series of tacks and measure how much time is needed to complete a race track. Race track contains 3 legs:

cross wind: can do several legs between #1 and #2 to get an average sailing speed in cross winds. Doing several legs also helps to correct for any current effect. 

upwind: define the breadth of the upwind zone based on the experience from 0. : how long does it takes to tack? how much distance do you need to reach a good speed again? take at least 3 times this distance.

downwind: when going back from buoy 3 to 1/2.

Repeat the race track with only bow bending and only stern bending.

Repeat with the boom and different ballasts: none, 5x weight, 10x weight

For all tests, follow Protei on another sailing vessel to RC it.

Wind measurement: measure as often as possible the wind speed and direction from the mother sailing vessel. 

Camera and GPS: make sure they are time synchrone.

Also needed here: a clever way to measure heel angle: need to measure the heigh of the mast from a fixed point (for example water level). An altimeter on top of the mast? Then we need to perform a dry test to calibrate the heeling angle measure.

videos of track,

time it takes to complete the track

  • anemometer
  • GPS logger (Protei speed and course)
  • stopwatch
  • 3 buoys
  • motor boat
  • camera with video or time laps
  • tide currents prevision as a function of time
  • detailed wind forecast, optimally with speed and direction 

</Summer testing campaign  - August 2011>

Angle measurement